(L&N) Q&A Thread - Quick questions go here [2010-2019]

[Tanni]

Syllables consist of onset, nucleus and coda, right? Is there a special linguistic generic term/superordinated concept for all three of them? If not, is it acceptable/good practice to call them ''syllable parts'' or ''syllable components''?

How are clusters of consonants and vowels connecting words and compound words called?

So if cvc means any syllable, cvc+cvc = cvccvc, where the two cvc parts need not being equal, how is the cc-part called? If either cvc part is a word, how is the cc-part called? (The words need not be just one syllable, but could also be compounds. Here, I use cvc-structure only to be able to point out the cc-part. Even the c-parts might in fact be clusters of more than one consonant.) If you have cv+vc=cvvc, how is the vv-part called? So, if one word ends in a vowel cluster and another word begins with a vowel cluster (regardless if these are ''trivial'' clusters of length one), is there a special term for the resulting vowel cluster?

Nucleus+coda is called rhyme, is there a term for onset+nucleus?

[Maximillian]

Syllables consist of onset, nucleus and coda, right?

Right.

Is there a special linguistic generic term/superordinated concept for all three of them?

Syllable?

If not, is it acceptable/good practice to call them ''syllable parts'' or ''syllable components''?

Why not.

How are clusters of consonants and vowels connecting words and compound words called?

I don't think there is a name for it. You can use the term "word-boundary" or "root-boundary".

Nucleus+coda is called rhyme, is there a term for onset+nucleus?

No.

[Tanni]

Thanks for the quick answer!

Is there a special linguistic generic term/superordinated concept for all three of them?

Syllable?

''Syllable'' would be the parts put together. What I meant was a term for the collection of the parts of a syllable/syllables.

How are clusters of consonants and vowels connecting words and compound words called?

I don't think there is a name for it. You can use the term "word-boundary" or "root-boundary".

Word-boundary or root boundary would only refer to the border line between the words or roots respectively, not for the consonant cluster formed by the last consonant(s) of the first and the first consonant(s) of the second word. (The same with vowels.) I could call it ''middle cluster'', but if you have complex compounds, there will be seveal such clusters on several root or word boundaries.

[sangi39]

I'd say:

V-C^n.C^n-V = syllable boundary
V-C^n-V = intervocalic consonant cluster
V^n = vowel cluster or diphthong or monophthong or long vowel depending on the realisation of the original vowels when put together. More or less up to the writer.
V^n-V^n = syllable boundary (for example we might have C_/a.i/_C which becomees a diphthong C_[ai]_C )

That is to say that the series of sounds as a whole is an intervocaic consonant cluster but the point where the first syllable ends and the second begins is called a syllable boundary.

As for a collective term for onset, nucleus and coda we could just call them "constituent parts" of the syllable. There's no single linguistic term for them as far as I know.

[Tanni]

Thanks, sangi39!

V-C^n.C^n-V = syllable boundary
V-C^n-V = intervocalic consonant cluster
V^n = vowel cluster or diphthong or monophthong or long vowel depending on the realisation of the original vowels when put together. More or less up to the writer.
V^n-V^n = syllable boundary (for example we might have C_/a.i/_C which becomees a diphthong C_[ai]_C )

Does n include 1?

I asked because I need names for variables etc. for some programs doing some text analysis. So the names shouldn't be too long, especially because they will add parts like ''_string'' or ''_stringsize''. So there could be a variable ''cluster'' of type ''clusterstring'', which itself is a string of ''clusterstringsize'' maximum length. So intervocalic_consonant_cluster seems a bit long to me. And I try do aviod abbrivations because of lack of readability. But if their're to long, they aren't readable, too. I thought about ''c_run'' and ''v_run'' for consonant run and vowel run, though.

That is to say that the series of sounds as a whole is an intervocaic consonant cluster but the point where the first syllable ends and the second begins is called a syllable boundary.

As for a collective term for onset, nucleus and coda we could just call them "constituent parts" of the syllable. There's no single linguistic term for them as far as I know.

There are obviously two kinds of intervocalic consonant clusters: within a simple word: the ''v'' in ''give'' or the ''ng'' in ''sangi''; and within the cluster on a word boudary, see the ''nl'' ''conlang'', even if ''con'' isn't a word here, and ''lang'' is just an abbreivation for ''language'', but you know what I mean. ''Constituent parts of the syllable'' is rather long for my purposes! But maybe ''constituent'' will do.

[Micamo]

How do phonological constraints evolve? Say the bilabial trill appears in only one root. It's obvious this consonant won't stick around for long. It'll probably merge with another consonant in the inventory. But what exactly is the process that determines which one it merges with?

[jseamus]

Optimality Theory (OT) can come in handy. Abridged Version: OT views constraints as not merely bivalent (present or absent), but as ordered on a priority hierarchy ascending from the most violable constraints to the least. Over time, a particular constraint may change its place on a particular languages constraint hierarchy, thus creating a new total phonotactic system.

For example, imagine that right now you have a constraint (RestrictCoda) that says that codas can only be nasals, laterals, or voiceless plosives, but it's dominated by another constraint (CodaIdentity) that says that codas must remain faithful to the input. Then RestrictCoda moves up in the hierarchy, dominating the CodaIdentity.

Now whether a coda has allowed or disallowed phonemes is more critical than whether it is faithful to the input. The output is much more likely to have only the allowed phonemes in the coda now. (See strict domination in the Wikipedia article).

BTW, why is /ʙ/ in the phoneme inventory at all?

[Micamo]

Optimality Theory (OT) can come in handy. Abridged Version: OT views constraints as not merely bivalent (present or absent), but as ordered on a priority hierarchy ascending from the most violable constraints to the least. Over time, a particular constraint may change its place on a particular languages constraint hierarchy, thus creating a new total phonotactic system.

For example, imagine that right now you have a constraint (RestrictCoda) that says that codas can only be nasals, laterals, or voiceless plosives, but it's dominated by another constraint (CodaIdentity) that says that codas must remain faithful to the input. Then RestrictCoda moves up in the hierarchy, dominating the CodaIdentity.

Now whether a coda has allowed or disallowed phonemes is more critical than whether it is faithful to the input. The output is much more likely to have only the allowed phonemes in the coda now. (See strict domination in the Wikipedia article).

Very interesting stuff. Thanks! I don't know if this answers my question or not yet but I'll be sure to give this a look over either way.

BTW, why is /ʙ/ in the phoneme inventory at all?

Remember the congenesis thread? [ʙ̩] is a root I came up with meaning "cold." It's supposed to imitate shivering.

[Ossicone]

BTW, why is /ʙ/ in the phoneme inventory at all?

Remember the congenesis thread? [ʙ̩] is a root I came up with meaning "cold." It's supposed to imitate shivering.

/ʙ/ is not that ridiculous sounding. I think you guys overemphasize it, so it become ridiculous.
When I was studying a lang with /ʙ/ half the time we though it was plain old /b/, except we could tell by looking at the speakers lips it was not.

[Ossicone]

Optimality Theory (OT)

OT is fun. Inyauk has ordered rules. (I will promote Inyauk till the day I die.)

How do phonological constraints evolve? Say the bilabial trill appears in only one root. It's obvious this consonant won't stick around for long. It'll probably merge with another consonant in the inventory. But what exactly is the process that determines which one it merges with?

I don't know. I don't really see a lang having a sound present in only one root.

However, if it did there a couple of options in my head:
1. The sound spreads itself to another phoneme and replaces it.
2. The sound spreads itself to another phoneme as an allophone. (Where it can live happily, become #1 or become #4)
3. Another sound spreads itself to the original sound and takes it as an allophone (Where it can live happily, become #1 or become #4.)
4. Another sound spreads itself to the original sound and replaces it.

So I guess there's two results, replacement or allophony, but multiple paths.

[Micamo]

I don't know. I don't really see a lang having a sound present in only one root.

However, if it did there a couple of options in my head:
1. The sound spreads itself to another phoneme and replaces it.
2. The sound spreads itself to another phoneme as an allophone. (Where it can live happily, become #1 or become #4)
3. Another sound spreads itself to the original sound and takes it as an allophone (Where it can live happily, become #1 or become #4.)
4. Another sound spreads itself to the original sound and replaces it.

So I guess there's two results, replacement or allophony, but multiple paths.

Well, the "sound in only one root" thing is a bit of an extreme example, but this general question can be applied to any phoneme merger. What, precisely, determines what it merges with? Say the lang with /ʙ/ also has /ɸ/ and /p/. For /ʙ/ to merge into either one seems, to me, about equally possible, you could make good arguments for similarity for either one. So, which one will it merge into? Possibly both depending on other conditions? Is it merely an arbitrary choice of the speakers or is there another process underlying this choice?

/ʙ/ is not that ridiculous sounding. I think you guys overemphasize it, so it become ridiculous.
When I was studying a lang with /ʙ/ half the time we though it was plain old /b/, except we could tell by looking at the speakers lips it was not.

Well, [ʙ̩ː] then.

[Ossicone]

I don't know. I don't really see a lang having a sound present in only one root.

However, if it did there a couple of options in my head:
1. The sound spreads itself to another phoneme and replaces it.
2. The sound spreads itself to another phoneme as an allophone. (Where it can live happily, become #1 or become #4)
3. Another sound spreads itself to the original sound and takes it as an allophone (Where it can live happily, become #1 or become #4.)
4. Another sound spreads itself to the original sound and replaces it.

So I guess there's two results, replacement or allophony, but multiple paths.

Well, the "sound in only one root" thing is a bit of an extreme example, but this general question can be applied to any phoneme merger. What, precisely, determines what it merges with? Say the lang with /ʙ/ also has /ɸ/ and /p/. For /ʙ/ to merge into either one seems, to me, about equally possible, you could make good arguments for similarity for either one. So, which one will it merge into? Possibly both depending on other conditions? Is it merely an arbitrary choice of the speakers or is there another process underlying this choice?

I think to some extent the choice is arbitrary.
Personally, I like using Distinctive Features. Here's a good link too.

So if you have:

Code: Select all

   ʙ    |     p    |    b    |    ɸ
+ cons    + cons     + cons    + cons
+ sonor   - sonor    - sonor   - sonor

+ voice   - voice    + voice   - voice
+ cont    - cont     - cont    + cont

By changing the features (sonority, voicing or continuity) B could change into any of those. So it's arbitrary in which, but more likely one of those than t or l or something else.

There's more but I'm lazy.

[Micamo]

I think to some extent the choice is arbitrary.
Personally, I like using Distinctive Features. Here's a good link too.

So if you have:

Code: Select all

   ʙ    |     p    |    b    |    ɸ
+ cons    + cons     + cons    + cons
+ sonor   - sonor    - sonor   - sonor

+ voice   - voice    + voice   - voice
+ cont    - cont     - cont    + cont

By changing the features (sonority, voicing or continuity) B could change into any of those. So it's arbitrary in which, but more likely one of those than t or l or something else.

There's more but I'm lazy.

If, other than place, those are the only things that are distinguished, then it looks to me like it'd turn out to be an allophone of /m/. Though judging consonants by the distinctive features the language contrasts is a very simplified view, just because a language has phonemic nasals and velars doesn't mean it'll also have a velar nasal. Still, it's at least an interesting thing to consider. Thank you!

[Ossicone]

If, other than place, those are the only things that are distinguished, then it looks to me like it'd turn out to be an allophone of /m/. Though judging consonants by the distinctive features the language contrasts is a very simplified view, just because a language has phonemic nasals and velars doesn't mean it'll also have a velar nasal.

It's not about the features the language distinguishes, it's about what seperates the phonemes. A phoneme is more likely to morph into something it has something in common with.

b is not going to turn into ŋ
but it could go to p, m, ɸ, β, v, ʙ, w.

[Raydred]

[ɫ]

I'm a bit late but thanks for the info.

[Micamo]

How normal/rare is it for a language which undergoes productive consonant mutation to have consonant phonemes which only occur as a product of these mutations? For example, productive /v/ > /v\/ where /v\/ doesn't occur in unmodified root words.

[sangi39]

How normal/rare is it for a language which undergoes productive consonant mutation to have consonant phonemes which only occur as a product of these mutations? For example, productive /v/ > /v\/ where /v\/ doesn't occur in unmodified root words.

From what I've seen it appears to be more common for the mutated form of a given sound to be a sound which is also found in other environments. However, Welsh provides a nice counter-example in that the voiceless nasals, AFAICT, only appear in word-initial position as the result of nasal mutation. Similarly, it is believed that the Irish Gaelic lenition of /mˠ/>/w/ and /mʲ/>/vʲ/, whereby the nasal becomes a fricative/approximant, originally came from the nasal leniting to a nasalised fricative which, AFAICT again, only appeared as the result of initial consonant mutation*.

Generally speaking then, it seems more common for a mutated form to be the same as a pre-existing phoneme which appears outside of the context of consonant mutation but it is possible for them to be distinct sounds in this particular context as well :) I'd additionally add, using the example from Irish Gaelic that sounds found in this specific context appear likely to merge further down the line with sounds which appear outside of consonant mutation. Welsh speakers, apparently, are more likely to pronounce the voiceless nasals as sequences of voiced nasals, pronounced as the final sound of the word causing the mutation, followed by an /h/, pronounced as the initial sound of the word undergoing mutation, thus "yng Nghaerdydd" is pronounced as if it were "yng Haerdydd" and "fy nhadau" as if it were "fyn hadau". This seems to further the idea of mergers between existing sounds but in this case the mutated form is split, leading to a retention between the mutated forms of voiced and voiceless plosives)

*Saying that, though, Modern Irish orthgraphy would indicate that it also appeared post/intervocalically as well, and considering lenition is common in these positions it would make sense for this nasalised fricative to appear here too, so it might have been a phoneme in other environments as well as initially.

[Micamo]

I thought so, but thanks for the confirmation. Interesting example with Irish Gaelic though.

[sangi39]

I thought so, but thanks for the confirmation. Interesting example with Irish Gaelic though.

S'alright. And if you haven't noticed (I made the edit after you posted, I think) I made and edit regarding the current pronunciation of voiceless nasals amongst at least some Welsh speakers which seems to follow the trend of merging mutation-only forms with pre-existing phonemes.

[teh_Foxx0rz]

How do different languages handle the non-questioning kind of which, as in "(the thing) which..." or "it does this, which..." (I know there's a word for it.)? What are the ways this can be handled?

And also, how can the same kind of "when" be handled, as in "such-and-such was happening, when..."?